On 10/9/25 12:10 PM, Luca Weiss wrote: > On Thu Oct 9, 2025 at 11:22 AM CEST, Konrad Dybcio wrote: >> On 10/9/25 11:16 AM, Luca Weiss wrote: >>> Hi Konrad, >>> >>> On Wed Oct 1, 2025 at 10:30 AM CEST, Konrad Dybcio wrote: >>>> On 9/30/25 3:57 PM, Luca Weiss wrote: >>>>> Describe yet another regulator-fixed on this board, powering the ToF >>>>> sensor. >>>>> >>>>> Signed-off-by: Luca Weiss <[email protected]> >>>>> --- >>>>> arch/arm64/boot/dts/qcom/qcm6490-fairphone-fp5.dts | 13 +++++++++++++ >>>>> 1 file changed, 13 insertions(+) >>>>> >>>>> diff --git a/arch/arm64/boot/dts/qcom/qcm6490-fairphone-fp5.dts >>>>> b/arch/arm64/boot/dts/qcom/qcm6490-fairphone-fp5.dts >>>>> index >>>>> 36d5750584831d66b4c2faf6042e4cbb3274eca7..0a64e5721e092d1f3e4bb7329335704eee567761 >>>>> 100644 >>>>> --- a/arch/arm64/boot/dts/qcom/qcm6490-fairphone-fp5.dts >>>>> +++ b/arch/arm64/boot/dts/qcom/qcm6490-fairphone-fp5.dts >>>>> @@ -195,6 +195,19 @@ vreg_usb_redrive_1v8: regulator-usb-redrive-1v8 { >>>>> pinctrl-names = "default"; >>>>> }; >>>>> >>>>> + vreg_vtof_ldo_2p8: regulator-vtof-ldo-2p8 { >>>>> + compatible = "regulator-fixed"; >>>>> + regulator-name = "VTOF_LDO_2P8"; >>>>> + regulator-min-microvolt = <2800000>; >>>>> + regulator-max-microvolt = <2800000>; >>>>> + regulator-enable-ramp-delay = <233>; >>>>> + >>>>> + gpio = <&tlmm 141 GPIO_ACTIVE_HIGH>; >>>> >>>> You may want to define the pincfg/mux config for this gpio too >>> >>> While I wouldn't say it's not good to have it, there's plenty of GPIOs >>> that have no pinctrl for it. Downstream doesn't set anything for gpio141 >>> either. >>> >>> I honestly wouldn't even know what the 'default' for a GPIO is in the >>> first place, or could I query the runtime state from the kernel? Is >>> /sys/kernel/debug/pinctrl/f100000.pinctrl/pinconf-groups trustworthy to >>> solidify this in the dts? >> >> I normally use /sys/kernel/debug/gpios > > Oh, if that works at least the path is a bit more memorable. I needed to > check quite some files in this directory to find the correct one. > >> >>> >>> 141 (gpio141): input bias disabled, output drive strength (2 mA), output >>> enabled, pin output (0 level) >> >> but this seems to be formatted very similarly if not identically >> >> Generally it reads out HW state, via (among other things) >> msm_config_group_get() > > So, you recommend setting a pinctrl for every single GPIO that's > referenced in the dts? Shall I send a patch to add all the missing ones?
That would make the state always deterministic and it's probably desirable yeah Konrad
